Scholarlink

<p>Software-defined networking (SDN) offers centralized control and programmability, but faces challenges in real-time adaptability and autonomous resilience against cyber threats and performance bottlenecks. This paper introduces a novel bio-inspired AI-driven self-healing framework for SDNs that integrates swarm intelligence for anomaly detection, genetic algorithms for network reconfiguration, and reinforcement learning for adaptive traffic optimization. Experimental evaluation using Mininet and the ONOS controller demonstrates that the proposed framework significantly outperforms traditional SDN mechanisms, achieving a 32% enhancement in attack mitigation efficiency, a 47% reduction in network recovery time, and a 29% increase in overall traffic throughput. These results underscore the potential of bio-inspired AI to enable autonomous, resilient, and high-performance networks for next-generation applications in cloud data centers and IoT.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Bio-inspired AI for adaptive and resilient software-defined networks: a self-healing approach for autonomous optimization and security

  • 期刊论文

  • 发表日期：

    2025年12月29日

查看全文
  • Soha Rawas,
  • Agariadne Dwinggo Samala

摘要

Software-defined networking (SDN) offers centralized control and programmability, but faces challenges in real-time adaptability and autonomous resilience against cyber threats and performance bottlenecks. This paper introduces a novel bio-inspired AI-driven self-healing framework for SDNs that integrates swarm intelligence for anomaly detection, genetic algorithms for network reconfiguration, and reinforcement learning for adaptive traffic optimization. Experimental evaluation using Mininet and the ONOS controller demonstrates that the proposed framework significantly outperforms traditional SDN mechanisms, achieving a 32% enhancement in attack mitigation efficiency, a 47% reduction in network recovery time, and a 29% increase in overall traffic throughput. These results underscore the potential of bio-inspired AI to enable autonomous, resilient, and high-performance networks for next-generation applications in cloud data centers and IoT.